Production of Somatic Hybrid Plants through Protoplast Fusion in Citrus

By SHOZO KOBAYASHI and TOSHIFUMI OHGAWARA*

Akitsu Branch, Fruit Tree Research Station (Akitsu, Hiroshima, 729-24 Japan)

* Research and Development Division, Kikkoman Corporation (Noda, Chiba, 278 Japan)

Citrus is one of the most important fruit­bearing trees in the world. Polyembryony and sterility often cause serious problems in citrus breeding. Few or no zygotic seedlings are produced when polyembryonic cultivars are used as maternal parents, because nucellar em­bryos restrict and often abolish zygotic em­bryo development prior to seed maturation1

) .

Protoplast fusion provides an alternative way of producing hybrids for species which can not be crossbred . Many inter- and intra­generic plants have been created by this tech­nique2l . The application of this technique to citrus would be of great value for the im­provement of this kind of fruit tree.

Here, we .report the production of some somatic hybrid plants in the Rutaceae family, although some of the results have already been reported elsewheres,11 >.

Materials and methods

1) Plan t materials Nucellar calli were induced from Trovita

orange (Citrus sinensis Osb.) and from F. N. Washington navel orange (C. sinensis Osb. var. brasiliensis Tanaka) as described pre­viously5> . About 1 g of the calli were suspend­ed in a 40 ml liquid Murashige and Tucker (MT) 9 > medium supplemented with 5% sucrose and 10 mg/ l 6-benzylaminopurine (BA) without auxin in a 125 ml Erlenmeyer flask. The cultures ,vere maintained on a rotary shaker at 110 rpm, and kept at 25°C under 16 hr/day illumination with cool fluorescent light (2,000 lux ) . Serial transfer of callus ,vas done every 2 weeks. Seeds of Poncirus trif o-

liata, Hayashi satsuma mandarin ( C. unshiu Marc.), Troyer citrange and Murcott tangor were germinated in a pot containing Ver­miculite. Plants (nucellar seedlings) were grown under the same environmental con­ditions as described in the callus culture. About 10 fully expanded leaves were harvested from 2-month-old plants.

2) Protoplast isolation Prior to isolation of the protoplasts from

suspention-cultured cells, 2-week-old cells were transferred to a hormone-free MT liquid medium (denoted to MT basal medium). After subculture in the same medium for 2 weeks, the cells were collected and subjected to proto­plast isolation using the procedure described previously0>.

In the case of the nucellar seedlings, leaves were rinsed with 70% ethanol, immersed in a solution containing 0.5% sodium hypochlorite plus 0.1 % Tween 20 for 20 min and washed twice with sterile distilled water. The leaves were then cut into about 2 mm wide strips with a razor blade and floated on a pretreat­ment solution (pH 5.8) containing 1 mM MES, 0.6 M mannitol and 1/2 strength of MT macro elements for 1 hr. About 0.6 g leaves were incubated in a Petri dish with 8 ml enzyme solution (pH 5.8), which consisted of pre­treatment solution supplemented with 3 % Cellulase Onozuka R-10 and 0.3% Macerozyme R-10. The incubation was carried out at 25°C on a rotary shaker (45 rpm) for 16 hr. The cell and enzyme mixture was filtered through a nylon mesh (58 µm pore openings), washed twice with 0.6 M mannitol by centrifugation

182

at 110 X g for 2 min. T hen, protoplasts were resuspended in the same solution.

S) Protovlast fusion and citlture Protoplast fusion was carried out at 5 com­

binations, that is Trovita orange plus P. trifo­liata, Trovita orange plus Troyer citrange, Trovita orange plus Hayashi satsuma man­darin, F. N. Washington naval orange plus Hayashi satsuma mandar in, and F . N. Washington naval orange plus Murcott tangor. Protoplasts of two cultivars were ad justed to a density of 10° cells/ml, mixed together at an equa.l volume, and fused with the aid of polyethylene glycol (PEG) by the method of Uchimiya 15>. PEG was diluted with 0.6 M mannitol-50 mM CaC12

and removed by centrifugation at 150 x g for 5 min. Protoplasts were washed twice with 0.6 M mannitol, and once with MT basal me­dium containing 0.6 M sucrose by centrifuga­t ion at 100 x g for 2 min. These protoplasts (105 cells/ml) were cultured in 3 ml medium, which consisted of MT basal medium con­taining 0.6 M sucrose and 0.6 % agarose (Sea Plaque, LMT, Marine Colloids) in a Falcon Petri dish (60 x 15 mm). The plates were sealed with Parafilm and maintained under 16 hr/day illumination with cool fluorescent light (500 lux) at 25°C. After 25th day, 0.5 ml MT basal medium was added to the protoplast culture, and plates were transferred to under 3,000 lux light intensity.

4) Plant regeneration Green embryoids (0.5-1 mm diameter) de­

rived from protoplasts were transferred to MT basal medium containing 500 mg/Z malt extract, 40 mg/l adenine and 0.8 % agar. They developed into cotyledonary embryoids after about 1 month. Cotyledonary embryoids de­veloped into whole plants within 6 months of culture when transferred to a MT agar me­dium containing 1 mg/l gibberellic acid.

5) Observation of chromosome n·umber Ten root tips of regenerated plants pre­

treated with 8-hydroxy-quinoline (2 mM) for 20 hr at 10°C were fixed in a mixed solution

J ARQ Vol. 22, No. 3, 1988

of ethanol: acetic acid (3: 1) for 24 hr, and then stained with lacto-propionil orcein for 3 hr according to Oiyama 121.

6) Analysis of leaf oil and J)e?·oxidase isozyme

Leaf oils were extracted from leaves of parents and those of regenerated plants, and then analysed by a gas chromatography ac­cording to the method described previously•> . Analysis of peroxidase isozyme was carried out by isoelectric focusing as described pre­viouslyn.

7) Analysis of ribosomal RNA genes (rDNA)

DNAs were extracted from leaves of par­ents and those of regenerated plants accol'ding to the method of Rogel's et a1. 1:1>. DNAs wel'e subjected to restriction endonuclease diges­tion, and followed by agarose electrophoresis and blot-hybridization with biotin-labeled rDNA fragments as probe. rDNA fragments were pl'epal'ed from recombinant plasmid pRR217 (kindly pt·ovided by Dr. K. Oono) which contained the whole rRN A gene se­quences of rice1•11 , and then labeled with biotin using biotin-11-dUTP and nick-trans­lation reagent kit (Bethesda Res. Lab., USA ) .

Plate I. Nucellar callus induced from the ovule of Trovita orange

\o/ - ltdl ,,

Flower of the Culture of the ovule

c;tn,, ····""· B """" from .h. n ....

Nucellar callus induced from the ovule

Nucellar callus protoplasts

Enzyme

treatment

Polyethylene glycol

Homokaryocyte __..~

Leaf of the Poncirus trifoliata

Mesophyll protoplasts

Homokaryocyte

Heterokaryocyte

Culture of the fusion products in a medium containing 0.6M sucrose

Citrus ,' Hetero l karyocyte ......... Poncirus

U n o o U H Q n u u Unorganized cell masses Embryoid

t No division

Somatic hybrid plant

Fig. l. An outline of somatic hybridization between Trovita orange and Poncirtts trifoliata

183

184

Visualization of the probe-target DNA hybrid was carried out using streptavidin-alkaline phosphatase conjugate, NET (nitro blue tetrazolium) and BCIP (5-bromo-4-chloro-3-indolyl phosphate, p-toluidine salt) .

Results and discussion

The nucellar callus was induced from ovules of unpollinated ftowers5 > (Plate 1) . It has been known that the nucellar callus induced from polyembryonic cultivars has a high potential to regenerate embryos, eventually the whole plants10>. Taking this advantage, Vardi et a1.1 n and we·•> have been successful to regenerate the whole plants from protoplasts

C

hy

p

' ' ' ' 4 8 12 16 20 24 28

Re tention time ( min)

Fig. 2. Gas chromatograms of leaf oil of Trovita orange (c), somatic hybrid ( hy) and P. trif oliat(I ( p)

32

JARQ Vol. 22, No. 3, 1988

of nucellar callus of some citrus cultivars. Therefore, we used the nucellar callus a., a partner of somatic hybridization.

An outline of somatic hybridization is drawn in Fig. 1. Protoplasts of the nucellar callus (Plate 2A ) had an ability to divide, proliferate and develop to green embryoids in a MT basal medium containing 0.25 M

mannitoJal . However, in the cultural condition of this study, most of the protoplasts pro­duced unorganized cell masses, and only a few occasionally developed into embryoids. Under the same conditions, mesophyll protoplasts of nucellar seedlings (Plate 2B) never divided. After the fusion treatment, heterokaryons (Plate 2D) were easily distinguished micro­scopically from other cells because of the exis tence of a colorless part from the cultured cell partner and a green portion from the mesophyll partner. About 50 days after cul­turing, many white unorganized cell masses and green embryoids were formed in the Petri dishes (Plate 2E) . These embryoids developed into the whole plants (Plate 2F) . In the fo l­lowing combinations, Trovita orange plus P. trifoli ata, Trovita orange plus Troyer citrange, Trovita orange plus Hayashi satsuma man­darin, and F. N. Washington navel orange plus Murcott tangor, only hybrid cells de­veloped into embryoicls in the presence of high concentrations of sucrose. But in the com­bination of F. N. Washington navel orange plus Hayashi satsuma mandarin, embryoids were formed not only from hybrid cells, but also from navel orange protoplasts.

A plant obtained by the protoplast fusion between Tl'Ovita' orange and P. trifoliata had charactel'istics of both parents. Leaves were trifoliate like P. trifolicita,, and their size. thickness and smoothness resembled those of Trovita orange (Plate 2G). A chromosome number of 36 was counted in the root tip of the plants (Plate 2H ) . Both parents have a chromosome number 2n = l8. In view of the intermediate leaf morphology and chromosome number, this plant must be somatic hybrid ( amphidiploid ) .

The isozyme analysis and the restriction endonuclease analysis of rDNA have been

[

Plate 2. Somatic hybridization between Trovita orange and P. trifotiata

• • • • •

A : Freshly isolated protoplasts from Trovita orange nucellar callus, B : Mesophyll protopJasts of P. trifoliata, C : Adhered protoplast (c : Trovita orange, p: P. trifoliata) , D : J-leterokaryon, E : An embryoid derived from a heterokaryon, F : A plant regenerated from an embryoid, G : Leaf morphology of a somatic hybrid plant and parents (o: Trovita orange,

hy : somatic hybrid, p : P. trifoliata) , H : A somatic hybrid plant had 36 chromosomes.

185

0

F

186

0 hy p t

Plate 3. lsoelectric focusing profiles of peroxidase from roots of Trovita orange (o), somatic hybrid ( hy), P. trifoliata ( p) and Troyer citrange ( t) (a sexual hybrid of C. sinensis and P. trifoliata)

p

JARQ Vol. 22, No. 3, 1988

0 hy ?

-9.3 ::7.98 .1

-6.5 -6.0

Plate 4. Blot-hybridization of biolabeled-rDNA fragments to Eco RI digested total DNA from leaves of P. trifoliata (p), Trovita orange (o) and somatic hybrid (hy)

Numerals indicate kbp.

Plate 5. Somatic hybrid plants obtained in the following combinations A : Trovita orange plus Troyer c itrange, B : Trovita orange plus Hayashi satsuma mandarin, C : F. N. Washington navel orange plus Hayashi satsuma mandarin, D : F. N. Washington navel orange plus Murcott tangor.

employed for the identification of somatic hybrid3,10>. Thus, we employed the leaf oil-, isozyme- and rDNA-analysis in the regener­ated plant and parents for the further con­firmation of the somatic hybrid. Gas chroma­togram of leaf oil of the regenerated plant was different from those of both parents (Fig. 2) . The zymograms of peroxidase showed that the regenerated plant had the specific band of Trovita orange and that of P. trifoliata (Plate 3) . Among the restriction endonuclease test­ed, Eco RI was shown to be the best enzyme for discriminating between rDNA fragments of Trovita orange and those of P. trifoliata. Clear and specific rDNA fragments originat­ing from nuclear DNA of Trovita orange \Vere 6.0 and 6.5 kbp, while those of P. trifoli<ita were 7.9, 8.1, 8.9 and 9.3 kbp. The regenerated plant had all of these fragments (Plate 4). These resu Its indicated that the regenerated plant was somatic hybrid. In the other com­binations, the regenerated plants (Plate 5) were also confirmed to be somatic hybrid (amphidiploid) by the analysis of rDNA and the observation of chromosome number.

In conclusion, we produced some somatic hybrid plants in the Rutaceae family by proto­plast fusion. Such hybrid plants would be useful for the practical citrus breeding pro­grams.

Summary

Somatic hybrid plants were obtained by protoplast fusion in the Rutaceae family. Protoplasts isolated from nucellar calli and from leaves of nucellar seedlings were fused by the PEG method. The fusion products were cultured in a Murashige and Tucker medium containing 0.6 M sucrose. In this me­dium, some colonies developed into the whole plants through embryogenesis. Almost all of the plants were shown to be somatic hybrid, which were proven by the restriction endo­nuclease analysis of nuclear ribosomal DNA. The chromosome number of the hybrid plants was 36, which was the sum of the parents (2n = 18) .

187

References

1) Button, J. & Kochba, J.: Tissue culture in the citrus industry. In Plant cell, tissue and organ culture. eds. Reinert, J . & Bajaj, Y. P . S., Springer-Verlag, Berlin, 70-92 (1977).

2) Gleba, Y. Y. & Sytnik, K. M. : P1·otoplast fusion (genetic engineering in higher plants). In Monographs on theoretical and applied genetics 8. Springer-Vel'lag, Berlin ( 1984) .

3) Handley, L. W. et al.: Somatic hybrid plants between Lycopersion esculentum and Solanitm, lycopersicoides. Theor. Appl. Genet., 71, 691-697 (1986).

4) Kobayashi, S., Uchimiya, H. & Ikeda, I.: Plant regeneration from 'Trovita' orange protoplasts. Jpn. J. Breed., 33, 119-122 (1983).

5) Kobayashi, S., Ikeda, I. & Nakatani, M.: Induction of nucellar callus from orange ( Citrus sinensis Osb.) ovules, and uniformity of regenerated plants. Bttll. Fruit Tree Res. Sta., E5, 43-54 (1984) [In Japanese with English summary].

6) Kobayashi, S., Ikeda, I. & Uchimiya, H.: Conditions for hig·h frequency embryogenesis from orange (Citrus sinensis Osb.) proto­plasts. Plant Cell Tissue 01·gan Culture, 4, 249-259 (1985).

7) Kobayashi, S. : Uniformity of plants re­generated from 01·ange (Citrus sinensis Osb.) protoplasts. Theor. Appl. Genet., 74, 10-14 (1987).

8) Kobayashi, S. et al. : A somatic hybrid plant obtained by protoplast fusion between navel orange (Citrus sinensis) and satsmna mandarin ( C. itnshiu) . Plant Cell Tissue Org<tn Culture, 14, 63- 69 (1988).

9) Murashige, T. & Tucker, D. P.H. : Growth factor requirements of citrus tissue culture. in Proc. 1st Int. Citrus Symp. ed. Chapman, H. D., Univ. Calif. Riverside, 3, 1155-1161 (1969).

10) Murashige, T. : Plant propagation through tissue cultures. Ann. Rev. Plant Physiol., 25, 135-166 (1974).

11) Ohgawara, T. et al.: Somatic hybrid plants obtained by protoplast f usion between Citrus sinensis and Poncirus t1·ifoliatlt . Theor. Appl. Genet., 71, 1- 4 (1985) .

12) Oiyama, I.: A technique for chromosome observation in root tip cells of citrus. Bitll. Fruit T1·ee Res. Sta., D3, 1-7 (1981) [In Japanese with English summary] .

13) Rogers, S. 0. & Bendich, A. J.: Extraction

188

of DNA from milligram amounts of fresh, herbarium and mummified plant tissues. Plant Mol. Biol., 5, 69-76 (1985).

14) Takaiwa, F ., Oono, K. & Sugiura, M. : The complete nucleotide sequence of a rice 17s rDNA gene. Nuc. Acid Res., 12, 5441-5448 (1984).

15) Uchimiya, H.: Somatic hybridization be­tween male sterile Nicotiana tabacuni and N. glutinosa through protoplast fusion. Theor. Appl. Genet., 61, 69-72 (1982).

JARQ Vol. 22, No. 3, 1988

16) Uchimiya, H. et al.: Detection of two dif­ferent nuclear genomes in parasexual hybrids by 1·ibosomal RNA gene analysis. Theor. Appl. Genet., 64, 117-118 (1983).

17) VaTdi, A., Spiegel-Roy, P. & Galun, E. : Plant regeneration from citrus protoplasts; variability in methodological requirements among cultivars and species. Theor. Appl. Genet., 62, 171- 176 (1982).

(Received for publication, Mal'ch 28, 1988)